A law that offers a four-day week and is barely used tells you more about design than about appetite for shorter time. Belgium’s experience is often quoted as proof that workers do not really want four days. Read closely, it shows something different: most workers do not want the same hours squeezed into longer days, with no reduction in total time and no guarantee about how the day off will be protected.
That distinction — reduced versus compressed — is the whole story here.
What Belgium actually legislated
Belgium’s Labour Deal, passed in 2022, gave full-time private-sector employees a right to request their normal working week over four days instead of five. The hours did not fall. A worker on a 38- or 40-hour contract could ask to work those hours in four longer days, subject to employer agreement and a written arrangement, typically for a six-month renewable period. The same package included a right to disconnect for larger employers.
This is a compressed week: 100% of the time in 80% of the days, at 100% of the pay, with no commitment to 100-80-100 output redesign. Days get longer, often to nine and a half hours or more, before breaks and commuting. Childcare pick-up times do not move because the law changed. Trains do not run later. Energy at the end of a long day is not the same as energy spread over five shorter days.
So when uptake proved low in the first two years — widely reported as a small single-digit share of eligible workers, and a fraction of one percent using the full-time four-day regime in employer surveys reported in 2023 and 2024 — the lesson is not that Belgians rejected shorter time. It is that they were not offered shorter time.
Why uptake stayed low
Three design features predicted the result.
First, the benefit is ambiguous. A day off is valuable; four long days can feel like paying for it in fatigue Monday to Thursday. Workers with caring responsibilities, health conditions, or long commutes often find compressed days harder, not easier. That is consistent with the wider evidence on long shifts: schedule satisfaction can rise for some, while tiredness and sickness risks rise for others, especially as daily hours climb.
Second, the right was a right to request. Employers could refuse with reasons, arrangements had to be negotiated, and reverting could be bureaucratic. Where trust is low, employees do not ask for visible flexibility they fear will mark them out.
Third, nothing reduced workload. No stop-doing list or coverage redesign was required. The calendar changed; the work system did not. Successful pilots redesign work first, then remove time.
What this does not prove
Belgium does not test the model Day5Group and the major pilots research: 100% pay, 80% time, and a commitment to maintain output through redesign. The UK pilot of 2022, the US and Ireland pilots of the same year, and Iceland’s public-sector trials between 2015 and 2019 all reduced total hours, usually from 40 toward 32 to 36, and they reported falls in burnout and stress with output broadly maintained. Those are different interventions with different mechanisms. More time for sleep, care, and recovery is the mechanism; compression offers less of it.
It also does not prove that compressed weeks are useless. Four longer days suit some shift patterns, some remote workers who want a full weekday for appointments, and some services that already roster long days. The honest claim is narrower: compressed weeks tend to deliver weaker and more mixed health gains than reduced weeks, and they transfer the cost of the day off onto the working days themselves.
Lessons for employers and policymakers
If you are a policymaker, decide which problem you are solving. If it is overwork and burnout, hours must fall. If it is scheduling choice, a compressed option is a modest flexibility tool — but label it that way, so the public is not told a four-day week arrived when a longer-day week did.
If you are an employer, do not copy the Belgian design and expect UK-pilot results. Ask staff which they would value: fewer hours, longer days with an extra day off, or a nine-day fortnight. You will get different answers by age, caring load, and commute — which is why staggered and mixed designs often beat a single firm-wide pattern.
Finally, protect the day. A compressed day off that is routinely used for catch-up work is not a benefit; it is unpaid overtime with better branding. Track actual hours, not rostered hours, whichever model you choose.
Belgium gave us a useful negative control. When the offer is the same work in longer days, most people decline. That strengthens, rather than weakens, the case for testing real reductions properly.
Sources
- Belgian Federal Public Service Employment, Labour and Social Dialogue, Labour Deal measures (2022) — compressed four-day request right and right to disconnect.
- Autonomy and Alda, Going Public: Iceland’s Journey to a Shorter Working Week (2021) — reduced-hours trials, 40 to 35–36 hours.
- Autonomy, University of Cambridge and Boston College, UK Four-Day Week Pilot results (February 2023) — 61 organisations, reduced time, wellbeing and retention findings.
- 4 Day Week Global, US and Ireland pilot results (December 2022) — 33 companies, reduced time, staff continuation preference.
Day5Group is a consultancy that assists in the transition to a 4 day work week. More about Day5Group · All blog articles